© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 68448
    Prev
    Next
    INDEX
    Explanations

    comparisons

    np_max-act · gemini-2.0-flash

    This neuron primarily activates on subword fragments that result from hyphenation line‐breaks (e.g. the pieces “non,” “re,” “new,” “able” in “non-renewable”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     await
    -0.07
    .getLogin
    -0.07
    待
    -0.07
     grabbed
    -0.07
    ieval
    -0.07
     Victory
    -0.07
    -days
    -0.06
    .completed
    -0.06
    ojení
    -0.06
     Pla
    -0.06
    POSITIVE LOGITS
     Techn
    0.07
    اهای
    0.07
    	Double
    0.06
     informant
    0.06
    DidEnter
    0.06
     Large
    0.06
    ENSOR
    0.06
     Robbins
    0.06
    _REGISTRY
    0.06
    _RENDER
    0.06
    Activations Density 0.027%

    No Known Activations